Game Artificial Intelligence

Taken Fall 2025, not Summer 2025

Fun course with really enjoyable assignments! If you took VGD before, you will be familiar with Unity. This class is more scripting in C# compared to that class, and less about playing around with the game objects in the scene (that's already set up for you in the assignments).

Professor Wilson's lectures are really informative, but he speaks a bit slow and I had to watch the videos at 1.5x speed to get through all of them. There's a lot of videos (some 1-1.5 hours) and it would have been hard to get through otherwise. The textbooks are optional.

Regarding the assignments, I think I actually spent more time on the earlier ones compared to the ones later in the course. Note that you only get two valid submissions in Gradescope for each assignment. A submission is considered valid if you score at least a 70%. If you have a second valid submission for an assignment, you score is the average of your two attempts.

Filtering the nodes and edges properly in the grid lattice/graph network for assignments 1-2 took me a bit of time, although a number of helper functions are already provided for you which makes this a lot less hard than it could have been.

Assignment 3 is implementing A* search from the Millington book which isn't bad, but I recommend making sure your node lists and path recursion are all correct before submitting in Gradescope.

Assignment 4 walks you through step by step what you have to do, but there is a catch with the expanded geometry that isn't explained. You have to basically add a check to ensure the candidate polygons are not sharing multiple edges at once.

Assignments 5 and 6 are dodgeball themed. 5 walks you through the steps again, but it takes a fair amount of time to tune to ensure your throws are accurate, especially when going against the Minion. For 6 I wish we actually had to implement the finite state machines ourselves but the starter code is mostly functional. For the dodgeball competition (optional extra credit) you have to tweak it a bit more but for the assignment itself you just have to add a conditional/constructor to ensure the Minion doesn't cross the half court line before throwing in the arena which doesn't take much time at all to implement.

For assignment 7, you use fuzzy logic to guide a racecar along different racetracks with different properties (one is straight, another is really curvy, another has bends but not as extreme, etc.). This is a HW where you could spend a lot of time just fine tuning things which is a really fun experience. You are graded based on whether you can hit the target speed for the course while not falling off the track more than once and get to decide what enum types to implement (how far to turn the wheel based on the position of the car on the track, and speeding up/slowing down). I did run into a Gradescope issue on this assignment where the car was not falling off as much locally (specifically on the sweepers track) compared to Gradescope but Prof Wilson helped point me towards an implementation with more consistent behavior.

The final assignment is the only one without coding. You are asked to develop a PCG terrain with three different biomes/features. The main challenge is playing with the rectangle function to separate out the different biomes. As long as none of your features (like the mountains) clip, it's relatively straightforward.

The quizzes are sneaky difficult, and are a lot more involved compared to VGD. It's less so a regurgitation of the lectures but more practical in nature. They give you a gaming scenario and you have to answer using the concepts from the lecture. Sometimes it involves math, other times it's terminology related. You should do well enough on the assignments that the quiz grades don't matter as much, but it's worth noting that many questions are all-or-nothing. There will be a five point question where you either get full points or 0 based on a correct or incorrect response which can be unforgiving and add difficulty to the course. You can only take the quiz once, so take your time and don't rush.

Overall, I really recommend this class as an elective. The TAs are really nice and responsive on Ed Discussion if you have any questions. If you do the work, you should get an A.

Actually took Fall 2025 This was a very fun course! I especially enjoyed the racecar and dodgeball projects. The lectures are super informative and applicable. And the projects are very reasonable. There were some topics covered in the lectures which could have made interesting projects, so I kind of wish there were more projects? Or maybe they replace some of the more boring projects like grid lattice? Anyways, I had a great time in this class and it was cool how enthusiastic students were about it.

A little about me: I have a bachelor’s degree in CS and about 1.5 years of SWE experience.

Courses I took before this one: IIS, SDP, and VGD.

I think this course is overall good and pretty manageable in the summer. The grades are basically made up of 8 different projects (and quizzes). You don’t really need to take VGD before this course either.

It’s the kind of class where you get out what you put in, I think some people can get by with ~5 hours a week, others might put in closer to 20 depending on how much effort you want to invest.

The only frustrating part for me was that you only get 2 submission attempts per assignment on Gradescope, but other than that, this course is great, definitely recommend.

Full review here: https://the11d.wordpress.com/2025/06/04/my-thoughts-on-game-ai-omscs-review-7/

TLDR (courtesy of GPT): In my review of Georgia Tech's OMSCS Game AI course (CS 7632), I shared my positive experience taking it in Summer 2024. Coming in with a background in Java/Kotlin and Python but no prior experience with C# or Unity, I found the Unity-based assignments both visually engaging and educational for understanding AI in games. The course had eight assignments and weekly quizzes, with a manageable weekly workload of around 7.3 hours. Some assignments, like Grid Lattice and Race Truck, were especially challenging due to their complexity and the need for precise tuning. I highlighted the importance of thoroughly understanding the provided boilerplate code and offered advice for future students. Overall, I found the course practical, rewarding, and a great fit for the Interactive Intelligence specialization.

This was a fulfilling class that taught me a lot about different AI implementations in Unity. The Fuzzy Logic project in particular was my favorite. It really made me think about how to define rules in such a way that the vehicle was great at both a straight away and twisty-turny type environments. I do think taking VGD before this class would have helped slightly (knowing how to debug using raycasts / lines would have saved me so much time), but the students in ed did a great job asking questions. Those questions gave excellent me excellent ideas for implementation, so I recommend checking ed for each assignment while you work on it.

The quizzes are slightly tough, but the assignments are doable and fun. This is not a difficult class, but it's one you will learn from!

Don't believe the people on here that are saying this is an easy course. The assignments are necessarily that hard, but they take a lot of time to get through and figure out. If you are like a lot of us and trying to balance actual work, family, and school, it is a huge time sink.

I was pretty disappointed with this course overall because I really liked VGD and Jeff's passion, but this class was very theoretical, math-y, and felt more like an undergrad weed out class than a graduate course. I would've liked to have few and less time-consuming projects that we could be more creative in instead of implementing things Unity or UE already just provides a library for like path planning.

The quizzes are all or nothing and don’t accurately test what you learned in the lectures. I think this class can take some tips from Joyner’s HCI class where the lectures are short and sweet and you get quizzed immediately after the lecture video. This helps solidify the knowledge in the moment instead of watching hours of videos only to hope that you paid attention during that one minute that explained the formula you know have to apply to a 5 point question or get a zero. There is a slight curve, but it doesn't make up for the absurd loss of 50% on every quiz.

This course has extremely well-designed assignments. This is an artificial intelligence course centered around video games. Some of the comments seem frustrated that this course is not the Video Game Design and Programming Course - I would not fault this course for that. If you want to learn methodologies for developing better artificial intelligence agents for video games as well as computational geometry, path planning, and procedural content generation - then this is the course for you!

Tips: 1.) There will be no teaching of C#, but honestly coming from a Java and C++ background, I had no trouble with C# and Unity. 2.) A lot of assignments are about creativity rather than trying to implement a certain algorithm optimally. Step back and take a big picture view of the assignment. What pieces of information would be valuable to your AI agents, how do you track them, and what do you do with them. Often, you can develop very simple, but powerful solutions if you do this. 3.) This course reflects the real world when it comes to submitting assignments. Your grade for an assignment is the average of your two allowed submissions. Therefore, you must understand the assignment and develop test cases for your code BEFORE submitting. You cannot farm a gradescope autograder to test and develop your assignment like you may have done in other courses.

I ended with a solid A and thoroughly enjoyed the course. There is a massive amount of lecture material, but it's quite valuable and the professor is extremely passionate!

I would say overall this was a decent course. I'm not that big into videogames, but I probably would've liked it more if I was. I enjoyed doing most of the projects, especially the racecar and dodgeball projects. The last project was almost more art than coding and a lot of people liked it but I thought it was meh. The lecture material had a lot of info, and a decent amount of material was not covered in any of the projects. I thought some of the lectures were dry. The quizzes were definitely not a gimme you had to prepare for them and make sure you understood the lectures pretty well, but I thought the quizzes were good for reinforcing the material. I think there is a lot of great material in this course if you are into video games.